Enhancing reader confidence through dual-time imaging and alternative reconstruction algorithms with [18F]PSMA-1007 PET/CT imaging in local relapsed castration-sensitive prostate cancer?

BACKGROUND: The current national and international guidelines recommend the use of PSMA imaging in proven BCR including very low levels of PSA (< 0.5 ng/ml), which is associated with a certain rate of false-negative results in regular clinical care. To address this issue, research efforts have focused on improving acquisition protocols via the implementation of delayed images or alternative reconstruction algorithms. One of the mainstays of this approach is the use of standardized reporting systems in regular clinical care and research setting to enhance reliability and reproducibility compared with unstructured confidence assessment. Thus, we aimed to investigate the added benefit of delayed imaging and the utility of reconstruction algorithms for the discrimination of equivocal [18F]PSMA-1007 findings in prostate bed. RESULTS: This monocentric, retrospective study enrolled 36 biologically male patients who underwent dual-time contrast-enhanced [18F]PSMA-1007 PET/CT scan between October 2021 and February 2024 due to BCR at a tertiary referral hospital. Histopathology after salvage surgery was considered as the gold standard, while clinical, biochemical, and radiological follow-up served as composite reference standards after reviewing the follow-up information. The minimum follow-up per patient was 15 months. The retrospective reading of the early-phase images revealed equivocal PSMA findings, i.e. PSMA-RADS 3 A, in the prostate bed in 12/36 patients (33%), with a total median SUVmax of 5.1 (3.7–6.7). The favourable results with delayed [18F]PSMA-1007 imaging led to upgrading of reporting in 8 out of 12 patients (75%) from PSMA-RADS 3 A to PSMA-RADS 4/5. CONCLUSION: In conclusion, based upon the study results, we suggest the introduction of a new algorithm to enhance and streamline the imaging decision-process, which would prevent avoidable follow-up scans or further imaging modalities and spare economic resources. Yet, further large-scale studies are warranted to validate the additive effect of this algorithm in regular clinical care.